Download Mutual Inductance of Two Helical Coils — Theory, Calculation

Progress In Electromagnetics Research Symposium Proceedings
2741
Mutual Inductance of Two Helical Coils — Theory, Calculation,
Verification
Michal Frivaldsky, Pavol Spanik, Marek Piri, and Viliam Jaros
Department of Mechatronics and Electronics, Faculty of Electrical Engineering
University of Zilina, Slovak Republic
Abstract— The paper deals about the verification of calculation process of a mutual inductance
between two helical coils, whereby backward computation of voltages and currents of designed
wireless energy transfer system is serving for confirmation of proposed methodology.
Nowadays, there are many mathematical procedures how to compute mutual inductance between
two coils. The main approach of proposed paper is exact specification of the formula, which may
be used for the determination of the mutual inductance. After mathematical computation were
done the confirmation of validity of proposed formula for mutual inductance calculation is verified
with simulation as well as with experimental measurement on proposed test bench of wireless
energy system. In order to determine the accuracy of investigated computation procedure backward confirmation has been provided in the way of comparison of time-waveforms of input/output
variables (voltages/currents) from measurement experiment on proposed wireless power transfer
(WPT) system with calculated time-waveforms.
1. INTRODUCTION — GENERAL THEORY
One of the important parameter for wireless power transmission is the mutual inductance that tells
us about how much is the electromagnetic coupling between two cylindrical coils strong, when an
air is located between them, and when they are positioned without coaxial and angular deflection.
Mutual inductance between two coils originates from the magnetic flux, which is generated by
one of the coils. For the sake of simplicity, let imagine two coils with one turn, whereby such a
coil can be defined as a loop. Transmitting loop generates magnetic flux Φ1 and electromagnetic
force ε. These variables are result of time-varying current flowing through transmitter. A part of
generated flux Φ1 is enveloped by receiving coil. This flux is designated as Φ12 and is defined as
mutual flux. Mutual flux for given area of coil generates magnetic induction, which is given by
Biot-Savavarth-Laplace law [X]-[X].
Mutual magnetic flux Φ12 , which flows through the area of receiving coil S can be defined as
follows:
¶
µ Z I
Z
dl2 × r21
µ0
Φ12 =
dS1 I2
(1)
B2 dS1 =
3
4π S1 l2
r21
Sz
In the same way the magnetic flux Φ21 can be defined and as valid for the reverse flow (i.e.,
receiving coil act as source).
µ Z I
¶
Z
µ0
dl1 × r12
Φ21 =
B1 dS2 =
dS2 I1
(2)
3
4π S2 l1
r12
S1
Formulas (1) and (2) can be further simplified into formulas (3) and (4):
Φ12 = L12 I2
Φ21 = L21 I1
Consequently for the calculation of mutual inductances, next formulas are valid:
Z I
µ0
dl2 × r21
L12 =
dS1
3
4π S1 l2
r21
Z I
µ0
dl1 × r12
L21 =
dS2
3
4π S2 l1
r12
(3)
(4)
(5)
(6)
Inductances L12 and L21 are the same. This fact can be confirmed, when we express magnetic
induction B in the way of vector potential [X]. Based on this, it can be stated that L12 = L21 = M ,
whereby M is defined as mutual inductance, whose static formula is as follows:
M=
Φ21
Φ12
=
I1
I2
(7)
PIERS Proceedings, Prague, Czech Republic, July 6–9, 2015
2742
The meaning of several symbols and variables is interpreted on Fig. 1.
Figure 1: Mutual magnetic coupling of two closed loop.
2. COMPUTATION FORMULAS FOR DETERMINATION OF MUTUAL INDUCTANCE
For calculation of the magnetic flux for round coil use of vector’s potential A is better choice,
instead of magnetic induction B. Vector’s potential is defined by
B = rot A
(8)
Then magnetic flux can be defined by integration of the vector’s potential along circumference
of the current turn.
I
(9)
Φ = A · dl
l
Advantage of using vector’s potential is that, we can work with it’s as a scalar quantity, for
a coaxial arrangement of the round coils. Vector’s potential driven by thin current turn can be
written as
Z 2π
µ0 I
cos ϕ0
p
Aϕ (r, ϕ, z) =
R1
dϕ0
(10)
2 − 2rR cos ϕ0 + (z − z 0 )2
2
4π
r
+
R
0
1
1
For better explanation of (10), see [1]. Magnetic flux for two coaxial thin round coils can by
written as:

! 
z=z
Z 22 z 0Z=z12ÃZ 2π
0
cos ϕ
µ0 N1 N2

p
ΦC/C =
R1 R2 I ·
dϕ0 dz 0  dz
2
2
2 h1 h2
R1 + R2 − 2R1 R2 cos ϕ0 + (z − z 0 )2
0
z=z21
z 0 =z11
(11)
Mutual inductance for two coaxial, thin, round coil can be defined by substituting (11) to the (7)

! 
z=z
Z 22 z 0Z=z12ÃZ 2π
0
µ0 N1 N2
cos ϕ

p
MC/C =
R1 R2 ·
dϕ0 dz 0  dz
2
2
0
0
2
2 h1 h2
R1 + R2 − 2R1 R2 cos ϕ + (z − z )
0
z=z21
z 0 =z11
(12)
This expression can be rewritten by elliptical integrals as:
3 N1 N2
4
MC/C = µ0 (R1 R2 ) 2
· [X(k11 ) − X(k22 ) − X(k33 ) + X(k44 )]
3
h1 h2
where function X(k) is defined as:
·
µ 2
¶¸
1 1 − k2
3ρ − 4
3
ρk − 2
2
X(k) =
(K(k) − E(k)) +
E(k) − ρ(1 − k )Π
,k
k
k2
2
2
ρ−2
(13)
(14)
where K(k) is elliptical integral of first kind, E(k) is elliptical integral of second kind and Π(n, k)
is elliptical integral of third kind. Elliptical integrals and relevant modules won’t be described in
detail here, but the relevant information can be found in [2].
Progress In Electromagnetics Research Symposium Proceedings
2743
3. CALCULATION AND EXPERIMENTAL MEASUREMENT OF MUTUAL
INDUCTANCE
The computation algorithm for calculation of self-inductance and mutual inductance of two air
coils is written in Matlab/Simulink. This can be divided into three parts (Fig. 2). First part is
part, where we define main parameters of both coils, e.g., radius or height of coil. Main parameters
of coils were taken over from existing system for wireless energy transfer (Fig. 3). Second part is
calculation of self-inductance. At first, the program calculates module the elliptic integrals and
then next subsystem calculates self-inductance. Then in next subsystem mutual inductance by
using (13) is being calculated. The results from computation are listed in Table 2.
As was already mentioned, the main parameters of the coils were taken over from existing
system for wireless power transfer (Fig. 3), which is primarily designed for the purposes of mutual
inductance investigation. The transmitting and receiving coils were designed with helical geometry.
As high frequency generator the evaluation board EPC9003 has been used [6]. Its main specification
is the use of perspective GaN power transistor devices (eGaN EPC2010C) which are suitable for
very high frequency operation.
Table 1 shows characteristic parameters of proposed helical coils for wireless energy transfer.
These parameters have been computed based on the target application of wireless system, whose
main parameters are parasitic resistance R and quality factor Q. Quality factor is derived from other
coil’s parameter. Target application of future proposal of this system shall be wireless charging of
Figure 2: GUI of computation algorithm for mutual and self-inductance calculation of two helical coils.
(a)
(b)
Figure 3: (a) Experimental test bench of wireless power transfer, for the purposes of mutual inductance
investigation and (b) principal schematics of transmitter and receiver configuration.
PIERS Proceedings, Prague, Czech Republic, July 6–9, 2015
2744
e-vehicles, thus existing model serves as testing sample in reduced ratio. After coil construction,
the main parameters influencing power transfer have been measured (self-inductance, parasitic
resistance). These parameters are listed in next table.
For the investigation of mutual inductance between transmitter and reciever, it was necessary
to measure input current of transmitting coil ILp and voltage on the load UZ . This is due to fact,
that mutual inductance can not be measured directly. The formula for the computation is:
M=
Uz
ωILp
(15)
Consequently we have determined the relative error between measured and computed values of
mutual inductance.
Mmes − Mcal
∆[%] =
× 100
(16)
Mcal
4. BACKWARD CONFIRMATION OF MATHEMATICAL CALCULATION OF MUTUAL
INDUCTANCE
Due to future purposes, where we would like to continue in the design of WPT for electro-mobility
applications, the simulation designing with mathematical approach of mutual inductance determination acts as flexible solution. The advantages of sim-math prototyping for the target application
are fast achievement of results, more flexible system optimization and no need of physical system
disposal. Due to these facts, the reliable simulation models must be prepared. Here we would
like to confirm the mathematical determination of mutual inductance in the way of mathematical
computation of time waveforms of input and output variables (voltages/currents) of proposed WPT
system. The results from mathematical computation are consequently compared with simulation
results from OrCAD PSpice, as well as with experimental measurements on physical sample.
Table 1: Computed parameters of proposed coils for wireless power transfer.
Param.
D
l
a
N
p
lw
L
R
C
Q
Lp
Ls
Rp
Rs
Description
Diameter of a air coil
Coil length
Wire diameter
Number of turns
Pitch between turns
Physical length of used wire
Inductance
Effective serial AC resistance
Parasitic capacitance
Quality factor
Inductance
Inductance
Serial DC resistance
Serial DC resistance
Value
185
60
1.5
6
10
3486
9.34
0.062
1020
174
9.86
10.12
1.09
0.8
Unit
mm
mm
mm
mm
mm
uH
Ohm
pF
uH
uH
Ohm
Ohm
Table 2: Measured parameters on a WPT system for mutual inductance calculation.
dis [mm]
100
120
140
160
180
UZ [V]
26.6
23.8
19.8
15.8
13.2
ILp [A]
8.4
10.4
11.6
12.4
12.8
Mmea [µH]
1.75362
1.2673
0.94523
0.7056
0.571
Mcal [µH]
1.6247
1.1949
0.899
0.6891
0.537
∆ [%]
6.54
4.56
3.59
7.88
4.59
Progress In Electromagnetics Research Symposium Proceedings
2745
The computation of input/output variables of proposed WPT system was done with the use of
next formulas (17–20). Due to complexity and mutual dependency of these variables, the GUI in
Matlab has been developed (Fig. 4).
duC1
1
=
iL
dt
C1 1
duC2
1
=
iL
dt
C2 2
diL1
L2
M
L2 (R1 )
= −
uC1 +
uC2 −
iL
2
2
dt
L1 L2 − M
L1 L2 − M
L1 L2 − M 2 1
M (R2 + Rload )
L2
+
iL2 +
uIN
2
L1 L2 − M
L1 L2 − M 2
M
L1
M (R1 )
diL2
=
uC −
uC +
iL
dt
L1 L2 − M 2 1 L1 L2 − M 2 2 L1 L2 − M 2 1
L1 (R2 + Rload )
M
−
iL2 −
uIN
L1 L2 − M 2
L1 L2 − M 2
(17)
(18)
(19)
(20)
Next figures (Fig. 5–Fig. 8) show time waveforms of the input/output variables which were displayed based on the results from computation (17)–(20). The main parameter, which was changing,
is the distance between transmitting and receiving coil (100 mm, 160 mm), thus in principle the
value of mutual inductance has been changed. To confirm the accuracy of the data, we used the
comparison with the experimental measurements, where distance between coils was also changed
similarly like in the case of computation. Here we would like to note that for the time waveforms
from measurements (Fig. 6, Fig. 8), the phase shift between voltage and current is visible. This
phase shift is caused due to insufficiency of measuring equipment, more exactly due to low measuring frequency range of current probe (HAMEG HZ 56) which is up to 100 kHz (the switching
frequency of proposed WPT system is 300 kHz).
Figure 4: Matlab GUI for computation of input/output variables of WPT system.
PIERS Proceedings, Prague, Czech Republic, July 6–9, 2015
2746
(a)
(b)
Figure 5: (a) Voltage and current on transmitting coil and voltage and (b) current on receiving coil for the
distance of 100 mm (results from computation).
voltage
voltage
current
current
(a)
(b)
Figure 6: (a) Voltage and current on transmitting coil and (b) voltage and current on receiving coil for the
distance of 100 mm (results from experimental measurement).
(a)
(b)
Figure 7: (a) Voltage and current on transmitting coil and (b) voltage and current on receiving coil for the
distance of 160 mm (results from computation).
In next table, the comparisons of the magnitudes of previously investigated variables are listed
together with the value of relative error.
Based on that, we can say that the proposed determination of the mutual inductance and
also proposed computational algorithm of time-waveforms of variables for WPT system can be
considered as reliable tool for future purposes, which will be related with design of the concept of
WPT for electromobility applications. The designed procedure is very attractive from the speed
Progress In Electromagnetics Research Symposium Proceedings
2747
voltage
voltage
current
current
(a)
(b)
Figure 8: (a) Voltage and current on transmitting coil and (b) voltage and current on receiving coil for the
distance of 160 mm (results from experimental measurement).
Table 3: Comparisons of magnitude input/output variables of proposed WPT system.
d = 100 [mm]
d = 160 [mm]
U1 m
[V]
15
15
computation
I1 m
U2 m
[A]
[V]
5.2
14.8
6.3
7.5
Relative error [%] (d = 100 mm)
Relative error [%] (d = 160 mm)
15
15
I2 m
[A]
0.9
0.47
5.2
6.3
U1 m
[V]
15
15
14.8
7.5
0.9
0.47
measurement
I1 m
U2 m
[A]
[V]
5.25
17.2
6.9
8.7
15
15
5.25
6.9
17.2
8.7
I2 m
[A]
0.95
0.55
0.95
0.55
of design point of view. A lot of physical designing can be eliminated, thus optimization of WPT
system can be done quickly with the used of proposed procedure.
The only problem which must be solved in order to improve accuracy of the results, is exact investigation and consequent definition of parasites, which are frequency dependent, and are
influencing efficiency of WPT (parasitic resistances, capacitances of transmitter and receiver).
5. CONCLUSION
In this paper, we were forced to evaluate the accuracy of the computation of mutual inductance
between two coils, whose might be suited for wireless energy transfer. The algorithm for the
computation of mutual inductance was designed based on [1].
The received results show that measured values are in acceptable accordance with computed
results, whereby maximal relative error was 7.88%. The main process which probably influenced
the accuracy of measurement was existence of frequency and temperature dependent parasitic components, mainly ESR of capacitors (it was series connection of set of capacitors). Also backward
confirmation of mutual inductance determination was realized in the way of comparisons of the
input/output variables of proposed WPT system. Computed and measured values were compared,
whereby the highest relative error was 17.2%. The main parameter, which influences these differences, is temperature and frequency dependency of parasitic elements. These facts must be included
into considerations in order to improve received results.
Future work will be focused on investigation of parasitic components, which are influencing
mutual inductance. Optimization process for their suppression will be done.
ACKNOWLEDGMENT
The authors wish to thank to Slovak grant agency APVV for project No. APVV-0433-12 — Research and development of intelligent system for wireless energy transfer in electromobility application.
2748
PIERS Proceedings, Prague, Czech Republic, July 6–9, 2015
REFERENCES
1. Pankrac, V., “Calculating the self and mutual inductance of coaxial coils in air; Power inductors
(Part 3),” Vol. 12, Departmet of Elektromagnetic Field CVUT in Prague, 2010.
2. http://mathworld.wolfram.com/EllipticIntegraloftheThirdKind.html.
3. Tirpák, A., “Elektromagnetizmus,” Vol. 521, 303–320, 2. vyd. Bratislava, 715 s, 2004, ISBN
80-88780-26-8.
4. https://www.scribd.com/doc/243062024/Ing-Darila-Matej.
5. http://www.mathworks.com/help/symbolic/mfunlist.html.
6. “Developement board EPC 9003C quick start guide,” http://epc-co.com/epc/documents
/guides/EPC9003 qsg.pdf.
7. Kindl, V., T. Kavalir, R. Pechanek, B. Skala, and J. Sobra, “Key construction aspects of
resonant wireless low power transfer system,” ELEKTRO, 2014, 303–306, May 19–20, 2014.
8. Brandstetter, P., P. Chlebis, and P. Palacky, “Application of RBF network in rotor time
constant adaptation,” Elektronika IR Elektrotechnika, No. 7, 21–26, 2011.
9. Grman, L., M. Hrasko, and J. Kuchta, “Single phase PWM rectifier in traction application,”
Journal of Electrical Engineering-Elektrotechnicky Casopis, Vol. 62, No. 4, 206–212, Aug. 2011.
10. Ferkova, Z., M. Franka, and J. Kuchta, “Electromagnetic design of ironless permanent magnet
synchronous linear motor,” International Symposium on Power Electronics, Electrical Drives,
Automation and Motion (SPEEDAM), 721–726, Italy, Jun. 11–13, 2008.
11. Kovacova, I. and D. Kovac, “Inductive coupling of power converter’s EMC,” Acta Polytechnica
Hungarica, Vol. 6, No. 2, 41–53, 2009.
12. Radvan, R., B. Dobrucky, and M. Frivaldsky, “Modelling of HF 200 kHz transformers for hardand soft-switching application,” Elektronika IR Elektrotechnika, No. 4, 7–12, 2011.